Fluid Diffusion In Microporous Molecular Dynamics Simulation

The last few years have heralded a rapid growth in the use of computer simulation as a tool for investigating the properties of both fluids in bulk and in confinement. Interest in molecular scale simulation has also increased, catalysed by the exciting developments nanotechnology such as the production of nano-machines, and small scale storage devices. Transport phenomena in fluids include viscosity, thermal conduction and diffusivity, refer to a phenomenon that gives rise to a flow of momentum, energy and mass. These phenomena are important as they describe how a fluid relaxes back to equilibrium following application of a mechanical or thermal perturbation. And hence, the transport coefficients are also important data in the fields of chemical engineering, materials, biosensors, micro chemical reactors and dug delivery. Knowledge of the behaviour of the corresponding transport properties with thermodynamic state, and with external fields, is necessary for achieving the aim of a unified theory of fluids. The ability to compute the transport coefficients from molecular simulation is also advantageous in engineering applications where standard tools such as computational fluid dynamics rely on this information, which is not always accessible from experiments. Consequently, simulation of transport in fluids is a very active area of research.Transport in porous catalysts and absorbents occur mainly by diffusion, which is often rate determining. A detailed understanding of diffusion in porous solids is therefore of great importance for optimization of these processes. For instance, the self-diffusion coefficients of fluids at a specific concentration and specific temperature is a key parameter for designing processes such as tertiary-oil recovery, dense fluids cleaning of solid surfaces and the extraction of essential oils from vegetable matrices, etc. The major contributions of this work are as follows: 1. Molecular dynamic simulations have been used to compute the mutual diffusion coefficients of equimolar Ar-Kr binary mixture confined in MOF-5. It is found that the distinct diffusivity D_d has larger negative contribution to mutual diffusivity D₁₂ at 200K than at 300K. The larger negative distinct diffusivity D_d at 200K can be attributed to the strong localization of guest atoms in the adsorption sites within the MOF-5.2. Equilibrium molecular dynamics simulations have been used to calculate the self-diffusion coefficients of fluid methane confined in mica slit pore at different temperatures, densities and pore widths. On the other hand, based on the Chapman-Enskog theory and Heyes relationships, two simple correlation models which can describe the self-diffusion coefficient of fluid methane in mica slit pore are proposed as a function of the temperature, density and pore width. Both these two models are written in terms of Lennard-Jones (LJ) dimensionless variables, which are defined in terms of the LJ parameters s and e. These parameters are meaningful at molecular level. The validity of these models is evaluated by comparing with the calculated self-diffusion coefficient data at different state points. The average error of both these two models is usually less than 13%.3. A simple mathematic correlation model has been proposed for the self-diffusion coefficients (SDC) of Lennard-Jones (LJ) fluids as a function of the reduced temperature and reduced density in bulk, and validated on the basement of the simulation results and experimental data at various conditions. The model indicates that the reduced SDC is proportional to the 0.2 power of reduced temperature and inverse proportional to the reduced density. The fitted results are imposing a good reproduction of these data and the total AAD are usually less than 10%. On the other hand, a mathematical model for describing the SDC of LJ fluids confined in slit pore has been proposed by basing on the SDC of fluid methane confined in mica slit pore. The model indicates that the reduced SDC is proportional to the 0.2 power of the reduced pore width and proportional to the 0.2 power of reduced temperature, but inverse proportional to the reduced density.